The performance of inductive CGP algorithms in the rediscovery of known virulence genes in all 3 GBS reference genomes.

This rediscovery analysis applied all known GBS virulence genes by applying stratified n-fold cross-validations with of the entire set of S. agalactiae genes in A909, NEM316, and 2603V/R genomes serving as test-set in each fold.n: number of genes in the category.a. scpB was also included as immune evasion genes.b. Including both bca and rib; also included as immune evasion genes.c. bac was represented by less than two genes in the three reference genomes studied. No rediscovery experiment was performed.d. Including all genes from the cps-neu operon. e. cspA was also included as an invasin.